www.nature.com/scientificreports

OPEN Comparative analysis of rhizosphere soil physiochemical characteristics and microbial communities between rusty and healthy ginseng root Xingbo Bian1, Shengyuan Xiao1,2, Yan Zhao1, Yonghua Xu1, He Yang1 & Lianxue Zhang1*

Ginseng rusty root (GRR) symptom is one of the primary diseases of ginseng. There has been a problem of ginseng rusty root, leading to a severe decline in the quality of ginseng. To clarify the relationship between root symptoms of ginseng rust and soil, the physical and chemical properties, enzyme activity, community structure and microbial diversity of GRR and healthy ginseng (HG) rhizosphere soil were analyzed and compared. The pH and redox potential (Eh) of GRR soil decreased, and the contents of total phosphorus (TP), available phosphorus (AP), and available potassium (AK) decreased. The activity of catalase and phosphatase and invertase was lower than that of HG groups. Besides, the microbial community of GRR rhizosphere soil changes much, and its abundance and diversity are signifcantly reduced. The community structure of GRR rhizosphere soil also shows apparent diferences, and the samples of the HG group gathered together, and the samples of the GRR group were dispersed. In general, GRR was closely associated with decreases in soil pH and Eh; decreases in TP, AP, and AK; decreases in the activity of several enzymes. Additionally, it is strongly associated with an increase in pathogenic microorganisms such as Ilyonectria and a reduction of benefcial microorganisms such as Tremellomycetes Acidobacteria subgroup 6 and Gemmatimonadetes.

Ginseng (Panax ginseng Mayer) is an essential medicinal material in China. Because of the relatively high plant- ing benefts of ginseng and the limitation of planting soil, ginseng production areas are relatively concentrated. Moreover, in recent years, the scale of cultivated ginseng in the cutting forests has been shrinking; it is increas- ingly essential to produce high-quality ginseng. One of the signifcant factors afecting the ginseng quality is the various diseases that occur during the growth of ginseng. Rusty root symptom is one of the central diseases in cultivation and production. It produces reddish-brown spots on the periderm of ginseng roots. With the increase of cultivation years, the patch may gradually expand, which will lead to the decline of commodity-grade and ginseng quality. At present, the mechanism of rust root is not precise. Whether ginseng rusty root is an infectious disease or a non-infectious physiological disease is still controversial, so some scholars have distinguished ­it1. Previous studies have shown that American ginseng (Panax quinquefolius L.) rusty root is the defense mechanism of ginseng itself, which is due to the invasion of some fungi that stimulates the production of phenolic compounds in ­ginseng2,3. Besides, some research results suggest that the induction of rusty root by fungal complex is the cause of rust roots in ginseng­ 4. A variety of enzymes were used to hydrolyze plant structural materials and make ginseng roots rusty. Fe­ 3+, combined with pectinase, had a synergistic efect on the formation of rust roots in ginseng­ 5. Current studies suggest that ginseng rusty root symptom is related to soil acidity and alkalinity, water content, and metal element content, as well as the accumulation and oxidation of phenolic compounds in roots.

1College of Chinese Medicinal Materials, Jilin Agricultural University, Changchun 130118, Jilin Province, China. 2National and Local Joint Engineering Research Center for Ginseng Breeding and Development, Changchun 130118, China. *email: [email protected]

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 1 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 1. Chemical properties of soil. (A) Soil redox potential; (B) pH. HG: healthy ginseng soil groups; GRR: ginseng rusty root symptom soil groups. Letters indicate signifcant diferences among diferent soil samples by ANOVA at P ** < 0.01 versus HG group (n = 6 in each group).

Soil microorganism is the most active part of the soil, driving force of soil material transformation and nutri- ent cycling. It participates in various soil processes, such as decomposition of soil organic matter, the formation of humus, transformation, and circulation of soil nutrients. Terefore, the composition and diversity of soil com- munities are closely related to soil function and plant health. Microorganisms in the soil are ofen the leading cause of plant diseases­ 6,7. Tese previous studies focused on the soil enzyme activity and nutrient characteristics of diferent tree species and their relationship with the incidence index of GRR. Besides, some scholars explored the soil characteristics of several ginseng farms with GRR to explore the relationship between GRR and soil characteristics­ 8,9. No data, however, are available on healthy ginseng and ginseng of rusty root on the same land on microbial community structure and enzyme activity in the plant rhizosphere. In previous studies, the roots of GRR were collected, and fungi were isolated from the lesions to reveal the pathogen causing GRR4​. Based on this phenomenon, it is hypothesized that some microbial communities should be closely related to GRR. By comparing to the rhizosphere soil of HG, any changes in the microbial communi- ties of GRR may be explored. In this study, we investigated the diferences in nutrient content, enzyme activity, and other physical and chemical properties between GRR and HG soils. pH, redox potential, catalase, phosphatase, urease, invertase activities, and soil phenolic acids were measured. In addition, the 16S and ITS rRNA gene fragment obtained directly from soil samples were sequenced to study the microbial community structure and diversity in soil samples. Results Chemical properties of soil. Count the degree of disease of 100 ginsengs and calculate the rusty root index based on the formula (Table S1). Calculated by formula, the rusty root index is 0.7725, and this is a ginseng farm with very severe ginseng rusty root (Fig. S1). In soil pH measurement, we found that the ginseng planting of soil with dangerous rusty root phenomenon decreased signifcantly (Fig. 1A). In soil Eh measurements, GRR soils decreased compared considerably to soils of healthy ginseng (Fig. 1B). Te nutrient content of the soil was also analyzed. Te results showed that there was no signifcant difer- ence in organic matter (OM), total nitrogen (TN), and available nitrogen (AN) between GRR soil and HG soil (Fig. 2A–C). Interestingly, however, tests of total phosphorus (TP), available phosphorus (AP), and available potassium (AK) showed a signifcant reduction in GRR soil compared to HG soil (Fig. 2D–F).

Enzyme activity of soil. Catalase, invertase, urease, and phosphatase of the two groups were compared. Among them, the activity of invertase, catalase and phosphatase in group GRR were signifcantly lower than those in the soil of healthy ginseng (Fig. 3A, B, D). However, there was no signifcant diference in urease activity among the two groups (Fig. 3C).

Microbial diversity and richness in GRR and HG rhizosphere soils. Te GRR and HG’s microbial diversity was characterized by partial 16S and ITS rRNA gene sequencing obtained from DNA directly extracted from soil samples. Rarefaction curves of observed amplicon sequence variants (ASVs) for each sample were produced at the ASV level (Fig. S2). Te curves of most of the samples reached close to the plateau, indicating that our sequencing depth was able to capture most bacterial species of ginseng rhizosphere soil microbiota. Compared with the rhizosphere soil of HG, the alpha diversity (including Chao1, Simpson, and Shannon diver- sity) of bacteria and fungi community in rhizosphere soil of GRR decreased signifcantly (Fig. 4). In other words, compared with the healthy ginseng rhizosphere soil, the microbial richness of GRR rhizosphere soil decreased signifcantly, whether it was bacteria (P < 0.05) or fungi (P < 0.05). Besides, the comparison between the two

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 2 Vol:.(1234567890) www.nature.com/scientificreports/

Figure 2. Nutrient content of soil. (A) Soil organic matter; (B) total nitrogen; (C) total phosphorus; (D) available nitrogen; (E) available phosphorus; (F) available potassium. HG: healthy ginseng soil groups; GRR: ginseng rusty root symptom soil groups. Letters indicate signifcant diferences among diferent soil samples by ANOVA at P * < 0.05, P ** < 0.01 versus HG group (n = 6 in each group).

Figure 3. Soil enzyme activity. (A) Catalase activity; (B) invertase activity; (C) urease activity; (D) phosphatase activity. HG: healthy ginseng soil group; GRR: ginseng rusty root symptom-high soil group. Letters indicate signifcant diferences among diferent soil samples by ANOVA at P ** < 0.01 versus HG group (n = 6 in each group).

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 3 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 4. Microbial richness and diversity. (A) Bacteria; (B) fungi. HG: healthy ginseng soil group; GRR: ginseng rusty root symptom soil group. Letters indicate signifcant diferences among diferent soil samples by ANOVA at P * < 0.05, P ** < 0.01 versus HG group (n = 6 in each group).

Figure 5. Microbial principal coordinate analysis (PCoA). (A) Bacteria Jaccard distance; (B) bacteria Bray– Curtis distance; (C) fungi Jaccard distance; (D) fungi Bray–Curtis distance. HG: healthy ginseng soil group; GRR: ginseng rusty root symptom soil group (n = 6 in each group).

groups of samples on Simpson and Shannon index showed that the diversity of bacteria and fungi in the rhizos- phere soil of GRR also decreased signifcantly (P < 0.05). According to the principal coordinate analysis (PCoA) based on Jaccard and Bray–Curtis matrix, the micro- bial community structure of rhizosphere soil is signifcantly diferent in HG and GRR (Fig. 5). Analysis of HG and GRR rhizosphere soil bacteria revealed a signifcant separation between the Jaccard (Fig. 5A) and Bray Curtis

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 4 Vol:.(1234567890) www.nature.com/scientificreports/

Figure 6. Venn diagrams of ASVs among soil samples. (A) Bacteria; (B) fungi. HG: healthy ginseng soil group; GRR: ginseng rusty root symptom soil group.

(Fig. 5B) matrices, indicating substantial diferences in bacterial composition between the two groups. In the fungi analysis, except that two samples in the GRR group in Fig. 5D are close to HG, other samples also show apparent separation. Te above results show that the composition of the two groups of fungi is also diferent. Interestingly, based on the Jaccard or Bray–Curtis diference matrix, the samples of the HG group were highly clustered, while the samples of the GRR group were highly separated. In other words, the rhizosphere soil microbial community of HG is relatively stable, and the microorganism of GRR rhizosphere soil showed even diferences.

Species diference analysis and biomarkers of GRR and HG rhizosphere soil. By sequencing, we obtained 33,798 ASVs in bacteria and 3,225 ASVs in fungi. We carried out the Venn map analysis of identify the potential biomarkers (taxa with stable signifcant diferences between groups) of bacteria and fungi in two groups of soil samples. As shown in Fig. 6A, in bacteria, the number of ASVs present in all six samples in the GRR group was 578, and the number of ASVs present in all samples in the HG group was 1,389. Two groups of common ASVs are 334. Compared with HG rhizosphere soil, the total ASVs of GRR rhizosphere soil decreased, and the specifc ASVs increased. In fungi, the 67 ASVs are common to all the samples, 26 specifc ASVs in the GRR group, and 82 specifc ASVs in the HG group (Fig. 6B). Tus, there was also a large diference in fungi ASVs between the rhizosphere soils of GRR and HG. In this study, we used the abundance data of the top 50 genera of the average abundance of bacteria to draw a heatmap (Fig. 7A). Ten, we use the histogram of LDA efect value of biomarker species based on linear dis- criminant analysis (LDA) efect size (Lefse) analysis to show the bacterial taxon with a signifcant diference between groups, and Fig. 7B shows the 20 taxa with the most signifcant diferences between the two groups (All signifcantly diferent taxa are provided in Fig. S3). Compared with HG rhizosphere soil, the abundance of Proteobacteria, Cyanobacteria, Oxyphotobacteria, Alphaproteobacterial, Gammaproteobacterial, Rhizobiales, Xanthomonadales, and Sphingomonadales in GRR rhizosphere soil signifcantly increased. On the other hand, the Deltaproteobacteria; Chlorofexi; Myxococcales; Gemmatimonadetes; Gemmatimonadales; Gemmatimona- daceae; Polyangiaceae; TK10; Acidimicrobiia is more abundant in HG groups compared to GRR groups. Besides, the cladogram of intergroup diference taxa based on Lefse analysis is provided in Figure S4. In the fungi analysis, we also used the abundance data of the frst 50 genera of average abundance to make the heatmap (Fig. 8A). Te histogram of the LDA efect value of 20 biomarkers, with the most signifcant diference between the two groups, is shown in Fig. 8B. Among them, the 20 most signifcant taxa of the GRR group are Minimelanolocus, Eurotiomycetes, Chaetothyriales, Herpotrichiellaceae, Helotiales, Leotiomycetes, Plectosphaere- lla, Nectriaceae, Ilyonectria and Hypocreales. Te most apparent ten taxa in the HG group are Mortierellaceae, Mortierellomycota, Mortierellomycetes, Tremellomycetes, Filobasidiales, Piskurozymaceae, Solicoccozyma, and Mrakiaceae (all signifcantly diferent taxa are provided in Fig. S3). Te cladogram of the intergroup diferences in fungi is presented in Figure S4.

Relationship of soil physicochemical properties and enzyme activity to microbial abun- dance. In redundancy analysis (RDA), we used measured soil physicochemical properties and multiple enzyme activities as environmental variables to investigate their correlation with high relative abundance taxa in soil samples. Te analysis results show that bacteria and fungi in two rhizosphere soils are greatly infuenced by physicochemical properties and enzyme activities. Furthermore, RDA plots refect the relationship between several benefcial or pathogenic microorganisms and environmental variables. In the analysis results of bacteria, only Cyanobacteria and Oxyphotobacteria showed a positive correlation with Urease. Gemmatimonadaceae, Gemmatimonadetes and Gemmatimonadales are positively correlated with a variety of environmental variables. Proteobacteria, Alphaproteobacteria, Rhizobiales, Gammaproteobacteria and Deltaproteobacteria have not shown a strong correlation with various environmental variables (Fig. 9A). In fungi, Ilyonectria and Hypocreales were negatively correlated with various environmental variables. Mortierellaceae Mortierellomycota, Mortierel-

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 5 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 7. Bacterial abundance and biomarkers. (A) Heatmap of the 50 most abundant genera in soil samples; (B) histogram of LDA efect value of the histogram of LDA efect value of diferentially abundant taxa. HG: healthy ginseng soil group; GRR: ginseng rusty root symptom soil group.

Figure 8. Fungi abundance and biomarkers. (A) Heatmap of the 50 most abundant genera in soil samples; (B) histogram of LDA efect value of diferentially abundant taxa. HG: healthy ginseng soil group; GRR: ginseng rusty root symptom soil group.

lomycetes Mortierellales and Mortierella were positively related to the AN and OM. However, there is no strong correlation between Solicoccozyma, Tremellomycetes and Filobasidiales and environmental variables. Discussion In this study, GRR rhizosphere soil and HG rhizosphere soil were compared comprehensively and carefully. Most of the previous studies were based on soil samples from diferent ginseng farms in diferent regions­ 10. To reduce the infuence of unknown diferences, the soil samples collected are from the same ginseng farm with a high rusty

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 6 Vol:.(1234567890) www.nature.com/scientificreports/

Figure 9. Redundancy analysis (RDA) of soil samples; (A) RDA plots of bacteria in soil samples; (B) RDA plots of fungi in soil samples. HG: healthy ginseng soil group; GRR: ginseng rusty root symptom soil group.

root index. To study the relationship between ginseng rusty root and soil, this study compared the soil physical and chemical properties and nutrient content, enzyme activity, and microbial community. Tis study examined the soil physical and chemical properties, nutrient content, enzyme activity, and microbial community to study the relationship between ginseng rusty root and soil. Soil pH seriously afects crop growth and microbial community structure in the soil­ 11,12. Compared with healthy ginseng rhizosphere soil, the pH of GRR rhizosphere soil decreased signifcantly (Fig. 1A). Ginseng in northern China usually grows in acid soil, but the lower pH value may afect the healthy growth of ginseng. Te pH has a particular infuence on microbial community structure, diversity, and richness­ 13,14. Terefore, fungi or bacteria related to GRR may appear in the soil with pH changes. Tere are many redox systems in soil, such as oxygen systems, iron system, manganese system. Under certain conditions, each soil has its Eh value­ 15. In this study, Eh value in GRR soil decreased signifcantly compared with that in HG soil (Fig. 1B). Its higher water content may cause the change of the Eh value of GRR soil, or some chemical reactions may occur in the soil. For a long time, high soil moisture content has been considered one of the factors causing GRR, and the part of this rust root is deemed to be the product of some chemical ­reactions2. Te growth of ginseng has high requirements for various substances in the soil. Phosphorus is an essential component of nuclear and membrane structure in plants. Normal phosphorus levels can complete the healthy metabolism of protein in plants, stimulate the growth of plant roots, increase the absorption of mineral nutrients by rhizomes, to alleviate the damage of plant diseases­ 16,17. In our results, GRR rhizosphere soil phosphorus signif- cantly decreased compared with HG rhizosphere soil (Fig. 2C, E). Besides, potassium can afect the metabolism of plants. Furthermore, potassium can promote the development of the thick outer wall of epidermal cells, thus preventing the occurrence of diseases­ 16,18. As shown in Fig. 2F, the GRR rhizosphere soil had signifcantly less potassium than the HG rhizosphere soil. Tis result seems to imply that GRR is closely related to the potassium content of the soil. Overall, the decrease of some nutrients in GRR soil may afect the healthy growth of ginseng. Te GRR may be related to reducing plant disease resistance caused by the reduction of phosphorus and potas- sium in the soil. Soil enzymes are the products of residual decomposition of plants and animals, exudation of plant roots, and metabolism of soil microorganisms, which involve many critical biochemical processes in ­soil19. Soil enzyme activity is an essential index of soil fertility, quality, and ­health20. Te catalase, invertase, urease and phosphatase activities in rhizosphere soil of ginseng were measured in this experiment. Te results showed that except ure- ase, the activities of the other three enzymes in ginseng rhizosphere soil decreased signifcantly (Fig. 3). Urease catalyzes the hydrolysis of urea to carbon dioxide and ammonia, which plays an essential role in the nitrogen cycle in ­soil21. In our results, it did not show a signifcant correlation with GRR. Invertase is a ubiquitous enzyme in soil, which plays an essential role in the release of fructose and glucose from sucrose and provides a carbon source for the growth of soil ­microorganisms22. Te organic phosphorus in the soil can be absorbed and utilized by crops under the action of phosphatase. Terefore, soil phosphatase activity can be an essential indicator of the hydrolysis of soil phosphorus ­compounds23. Te decrease of phosphorus in GRR soil may be related to the reduction of the enzyme activity. Catalase can drive the decomposition and transformation of peroxides in soil, eliminating the adverse efects of peroxides on soil ­quality24. In other words, the decrease of catalase activity in GRR soil may make ginseng sufer more oxidative ­stress25. In brief, the reduction of these three enzymes in GRR soil indicated that the production of ginseng rusty root might be related to the decrease of carbon source and available phosphorus, and the content of peroxides in the soil.

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 7 Vol.:(0123456789) www.nature.com/scientificreports/

In this study, high-throughput sequencing technology was used to study the microbial diversity of GRR rhizosphere soil. Te alpha diversity analysis of GRR and HG rhizosphere soil, the Chao1 estimator­ 26, Shannon­ 27, and Simpson­ 28 diversity of ginseng rhizosphere soil with rusty root showed a signifcant decline (Fig. 4). Tere- fore, in this study, the richness and diversity of fungi or bacteria in the rhizosphere soil of GRR signifcantly decreased. Ten, we conducted the beta diversity analysis based on PCoA, Jaccard, and Bray–Curtis distance was estimated in this study (Fig. 5). Jaccard distance and Bray–Curtis distance are measures used to measure the diference in species composition of diferent soil samples. Tey can calculate the characteristics of diferent species composition of samples. Te focus of the two analyses is diferent: Jaccard only considers the presence or absence of species in the sample, while Bray–Curtis considers the presence or absence of species in the sample and considers the relative richness of diferent species. Whether fungi or bacteria, the GRR group, and HG group had visible separation, which indicated that GRR rhizosphere soil microorganism had signifcant changes compared with HG soil microorganism. It should be noted that six samples of the HG group showed high aggregation. In comparison, six samples of the GRR group were very dispersed, which indicated that the microbial structure of HG rhizosphere soil was very stable. In contrast, that of GRR rhizosphere soil might have microbial community disorder. Finally, we analyzed the species that caused these diferences. Using the ASVs abundance table to make Venn diagrams for community analysis shows the ASV level variation between the two groups (Fig. 6). To further compare the diferences in species composition between samples, we use the abundance data of the top 50 genera of average abundance to draw a heatmap to display the genera abundance distribution trend of each sample (Figs. 7A, 8A). In this study, LEfSe analysis was used to fnd the biomarkers between groups, and LDA efect histograms and cladograms were drawn for bacteria and fungi, respectively (Figs. 7B, 8B and Fig. S4). We found possible pathogens in the GRR rhizosphere soil. However, the other results of LEfSe analyse were dif- ferent from previous studies, which may be related to the sample collection site and the cultivation conditions. In terms of fungi, the biomarker’s analysis found Ilyonectria, which is thought to be the pathogen of GRR​ 29. Hypocreales is ofen reported to be related to pests’ pathogenicity or some ­bacteria30–32, and it can survive in harsher soil ­conditions33. Nectriaceae is also listed as a biomarker. It includes numerous important plant and human pathogens, and several species used extensively in industrial and commercial applications as biodegrad- ers and biocontrol agents­ 30. We found a possible pathogen Plectosphaerella, it has been reported to be associated with plant root diseases­ 34, but it has not yet been reported on ginseng disease. In HG rhizosphere soil, Mor- tierellales, Tremellomycetes orders predominated. Mortierella is a biomarker in HG rhizosphere soil, but rarely in GRR soil. Mortierella is the producer of polyunsaturated fatty acids­ 35, whether it is related to ginseng growth has not been reported. Tere are reports that Tremellomycetes could efciently act as opportunists that utilize the decomposition products provided by other ­microbes36. RDA reveals the correlation between fungal taxa and environmental variables (Fig. 9B). Ilyonectria was negatively correlated with several environmental variables, especially pH, catalase, and invertase. On the other hand, in the analysis of bacteria, more proteobacteria appeared in GRR rhizosphere soil. Although proteobacteria have been reported to dominate pesticide and herbicide contaminated ­soils37,38, it is generally one of the most abundant phyla in many soils and rhizosphere soils. Te reasons for the increase of proteobacteria in GRR rhizosphere soil need to be further explored. A signifcant increase in Proteobacteria ofen indicates the dysbiosis in the microbiota. Further, a higher abundance of Alphaproteobacterial (including Sphin- gomonadales) and Gammaproteobacterial (including Xanthomonadales) indicates that GRR rhizosphere soil may have stronger nitrifcation and the turnover of OM­ 39. Interestingly, anaerobic Deltaproteobacteria (including Myxococcales) is more abundant in HG rhizosphere ­soil40. Myxococcus is considered to be an essential and active predator that regulates bacterial communities in agricultural land­ 41. Cyanobacteria are the main constituent of biological bacteria and can survive in more extreme environments­ 42. Some cyanobacteria have increasing soil stability and moisture-holding ­capacity43,44. Teir abundance increased in GRR rhizosphere soil, confrming the deterioration of GRR rhizosphere soil conditions compared to HG. Chlorofexi (including TK10) is abundant in HG rhizosphere soil. It seems to play a useful role in the soil in providing the flamentous scafolding around which focs are formed feed on the debris from lysed bacterial cells. It can ferment carbohydrates and degrade other complex polymeric organic compounds to low molecular weight substrates to support their growth and that of diferent bacterial ­populations45. Also, Acidobacteria subgroup 6 and Gemmatimonadetes emerged as the keystone taxa in HG rhizosphere soils, and they may play an important ecological role by degrading polysac- charides of plant and fungal ­origin46,47. Tese diferences could be partially due to the diferent physicochemical characteristics of the two ­soils48. Fungal plant pathogens live in a specifc environment, and the GRR rhizosphere soil environment may be more suitable for the survival of pathogenic fungi such as Ilyonectria. Many studies have pointed out that ginseng, American ginseng (Panax quiquefolium L), (Panax notoginseng (Burkill) F. H. Chen ex C. H.), and other medicinal plants have rusty root symptoms­ 29,49. Furthermore, they all seem to be closely related to Ilyonectria. Te same results were found in our study. However, in Wang et al.50, Ilyonectria does not have a large abundance in GRR rhizosphere soil. Also, in Liu et al. research results, Ilyonec- tria is positively correlated with N and ­OM10. In our analysis, there is a negative correlation. Plectosphaerella, as A potential pathogen, also showed greater abundance in our results, which had not been found in the study of Liu et al. In agreement with previous studies, we also found large abundances of Proteobacteria, Gammapro- teobacterial and Rhizobiales in GRR rhizosphere soil. What is diferent is that we also found a large abundance of Cyanobacteria. Interestingly, our analysis of HG rhizosphere soil bacterial and fungal abundance showed signifcant diferences from previous studies. Tus, the microbial composition of the rhizosphere soil of ginseng is likely to be closely related to the planting site, cultivation method, fertilizer and pesticide application. In conclusion, pH and Eh in GRR soil decreased signifcantly, which may be related to the metabolism of the ginseng root system and soil aeration. GRR also showed signifcant changes in soil nutrient content, reduc- ing TP, TK, and AK, making ginseng absorb fewer nutrients and may afect the disease resistance of ginseng. Te decrease in the activities of many enzymes in GRR soil indicates that the metabolism of soil substances

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 8 Vol:.(1234567890) www.nature.com/scientificreports/

is greatly afected. Besides, compared with HG rhizosphere soil, microbial abundance and diversity of GRR decreased signifcantly, community structure also changed dramatically, and community disorder occurred. Finally, through Lefse analysis, the number of microorganisms in GRR rhizosphere soil that may be benefcial to the soil decreased, and the pathogenic bacteria (Ilyonectria) mentioned in previous studies were also found. Soil is a complex environment. Physical and chemical properties, nutrients, and microbial communities afect the growth of ginseng. RDA analysis of the correlation between diferent microorganisms and environmental factors will help improve ginseng planting soil in the future. Tis study mainly studies the change of ginseng rusty root symptom soil, and ginseng rusty root symptom may be caused by many factors that need further investigation. Materials and methods Soil information and sample collection. All soil samples were collected from the same ginseng farm with ginseng (5-year-old) in Hunchun city, Jilin province, China (42.86′ N and 130.37′ E). In detail, this is a ginseng farm with an area of about 21,000 square meters, with an altitude of 40 m, belongs to the temperate marine climate, with an annual average temperature of 5.65 °C, average precipitation of 617.9 mm, a frost-free period of 140–160 days, and an average temperature of 21.2 °C in August. Te farm was the frst time to cultivate ginseng and was cultivating transplanted 2-year-old ginseng seedlings. Te pesticides applied in ginseng cultiva- tion are “carbendazim (systemic broad-spectrum fungicide)” bought from Shandong Xinnongji Biotechnology Co. LTD (China), “tiandashenbao (plant cell membrane stationary agent)” bought from Shandong Tianda Bio- pharmaceutical Co. LTD (China) and “metalaxyl (Phenylamide fungicides)” bought from Alta Technologies LTD (China). All pesticides are used strictly following the “Ginseng safe production technical specifcation of pesticide application (DB22/T 1233-2019)”. We randomly collected 200 ginseng and calculated the rusty root ­index9,51 to evaluate the severity of the disease.

n × rusty root grade Rusty root index = N × the highest rusty root grade

In this formula: n = Te quantity of ginseng in diferent rusty root grades, N = Total number of ginsengs used for statistics. Rusty root grade was divided into 0–4, which were healthy ginseng root, rust area < 10%, 10–25%, 25–50%, and > 50% in turn. Ten, six plots with ginseng grade 0 and six plots with ginseng grade 4 were selected to collect soil samples, about 1 square meter per plot. Te soils were sampled from each plot at depths of 5–15 cm, representing soils from the root zone and below the root zone. Te 12 soil samples are divided into the HG group (healthy gin- seng group) and GRR group (ginseng rusty root symptom group). Part of each soil sample is kept wet for the determination of soil Eh. Finally, six HG rhizosphere soils and six GRR (grade 4) rhizosphere soils were collected from the 12 plots above (one rhizosphere soil sample was collected at each polt). Part of the soil microbial DNA was extracted, and the samples were kept in the sterile centrifuge tube at—80 °C until use. Te other part of the soil was air-dried at ambient temperature, crushed, and sieved through a 2 mm plastic mesh to determine soil pH, soil nutrition, and soil enzyme activity­ 52. In this part, only the soil attached to the ginseng root is considered as rhizosphere soil­ 53.

Analysis of basic physical and chemical properties. Te pH value of soil was measured with a pH meter/potentiometer under the soil: water ratio of 1:2.5. Te Eh was determined by potentiometry (China HJ 746-2015)54. Te soil OM content was measured by the potassium dichromate external heating method. Te TN was determined by Kjeldahl ­method55, and the TP was determined by alkali fusion molybdenum antimony col- orimetry (China HJ 632-2011). Te AP was determined by ­NaHCO3 extraction molybdenum-antimony color- imetry; the AN was determined by the alkali difusion method; AK was determined with the fame photometric method­ 56.

Analysis of soil enzyme activity. Te activities of catalase, invertase, urease, and phosphatase in the soil of ginseng were measured. Te activity of catalase was determined by KMnO­ 4 titration; soil invertase activity was determined by 3,5-dinitrosalicylic acid colorimetry; urease activity was determined by indophenol blue colorimetry; acid phosphatase activity was determined by disodium phenyl phosphate colorimetry method­ 37.

DNA extraction and Illumina sequencing. DNA was extracted from 0.5 g of soil (wet weight) using the Mag-Bind soil DNA kit (Omega Bio-tek, Norcross, GA, U.S.) according to the manufacturer’s protocols. Ten, DNA was quantifed by Nanodrop (Termo Scientifc, Waltham, MA, U.S.) and the quality of DNA was detected by 1.2% agarose gel electrophoresis, the ­OD260/280 values of DNA extraction were between 1.8 and 2.0. Te V3-V4 region of bacterial 16s rRNA gene was amplifed by bacterial specifc primers 338F and 806R, and the internal transcribed spacer (ITS1) was amplifed by eukaryotic specifc primers ITS5F and ITS1R. In this step, the sample is uniformly diluted to 20 ng/ul. Amplifcation system (25 μL): 5 × reaction bufer 5 μL, 5 × GC bufer 5 μL, Dntp (2.5 Mm) 2 μL, Forwardprimer (10 uM) 1 μL, Reverseprimer (10 uM) 1 μL, DNA Template 2 μL, ddH­ 2O 8.75 μL, Q5 DNA Polymerase 0.25 μL (NEB). Amplifcation parameters: Initial denaturation 98 °C 2 min, Denaturation 98 °C 15 s, Annealing 55 °C 30 s, Extension 72 °C 30 s, Final extension 72 °C 5 min, 10 °C Hold. 25–30 Cycles. Te sequencing library was prepared by using Illumina’s TruSeq Nano DNA LT Library Prep Kit, the fnal fragment was selected and purifed by 2% agarose gel electrophoresis.

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 9 Vol.:(0123456789) www.nature.com/scientificreports/

Te library was quantifed by Quant-iT PicoGreen dsDNA Assay Kit in Promega QuantiFluor fuorescence quantitative system, and then libraries were sequenced on the Illumina HiseqPE250 platform at Personal Bio- technology Co., Ltd. (Shanghai, China).

Bioinformatics and data analysis. Microbiome bioinformatics was performed with QIIME 2 2019.457 with slight modifcation according to the ofcial tutorials (https​://docs.qiime​2.org/2019.4/tutor​ials/). Briefy, raw sequence data were demultiplexed using the demux plugin following by primers cutting with cutadapt ­plugin58. Sequences were then quality fltered, denoised, merged, and chimera removed using the DADA2 ­plugin59. Non-singleton ASVs were aligned with ­maf60 and used to construct a phylogeny with ­fasttree261. To analyze diferent samples using the same sequencing depth, all samples were re-sampled into the same sequence number: 119938 sequences from fungi and 91,828 sequences from bacteria. Alpha diversity metrics: Te Chao1 estimator, Simpson and Shannon diversity index. Beta diversity metrics: Jaccard distance and Bray–Curtis dis- similarity. Taxonomy was assigned to ASVs using the classify-sklearn naïve Bayes taxonomy classifer in feature- classifer ­plugin62. LEfSe was used to elucidate the biomarkers in each group­ 63. In addition, RDA was used to reveal the relationships between microbiota and soil properties. Tis analysis was carried out with the “vegan” package in R.

Statistical analyses. All data were analyzed using the SPSS sofware (IBM Corporation, Armonk, NY, USA), and the results were expressed as the arithmetic mean value ± standard deviation. Te diferences in the means were compared by the Student’s t test at P < 0.05.

Received: 4 April 2020; Accepted: 6 August 2020

References 1. Zhou, Y. et al. Changes in element accumulation, phenolic metabolism, and antioxidative enzyme activities in the red-skin roots of Panax ginseng. J. Ginseng Res. 41, 307–315 (2016). 2. Rahman, M. & Punja, Z. K. Biochemistry of ginseng root tissues afected by rusty root symptoms. Plant Physiol. Biochem. 43, 1103–1114 (2005). 3. Reeleder, R. D., Hoke, S. M. T. & Zhang, Y. Rusted root of ginseng (Panax quinquefolius) is caused by a species of rhexocer- cosporidium. Phytopathology 96, 1243–1254 (2006). 4. Lu, X. H. et al. Taxonomy of fungal complex causing red-skin root of Panax ginseng in China. J. Ginseng Res. 44, 506–518 (2020). 5. Lee, C., Kim, K. Y., Lee, J. E., Kim, S. & An, G. Enzymes hydrolyzing structural components and ferrous ion cause rusty-root symptom on ginseng (Panax ginseng). J. Microbiol. Biotechnol. 21, 192–196 (2011). 6. Yuan, X., Song, T. J., Yang, J. S., Huang, X. G. & Shi, J. Y. Changes of microbial community in the rhizosphere soil of Atractylodes macrocephala when encountering replant disease. S. Afr. J. Bot. 127, 129–135 (2019). 7. Mazzola, M. & Manici, L. M. Apple replant disease: role of microbial ecology in cause and control. Annu. Rev. Phytopathol. 50, 45–65 (2012). 8. Liu, X. et al. Comparison of the characteristics of artifcial ginseng bed soils in relation to the incidence of ginseng red skin disease. Exp. Agric. 50, 59–71 (2014). 9. Wang, Q. X. et al. Analysis of the relationship between rusty root incidences and soil properties in Panax ginseng. 41, 012001 (2016) 10. Liu, D., Sun, H. & Ma, H. Deciphering microbiome related to rusty roots of Panax ginseng and evaluation of antagonists against pathogenic ilyonectria. Front. Microbiol. 10, 1350 (2019). 11. Guo, J. H. et al. Signifcant acidifcation in major Chinese croplands. Science 327, 1008–1010 (2010). 12. Wan, W. et al. Responses of the rhizosphere bacterial community in acidic crop soil to pH: changes in diversity, composition, interaction, and function. Sci. Total Environ. 700, 134418 (2020). 13. Rousk, J. et al. Soil bacterial and fungal communities across a pH gradient in an arable soil. ISME J. 4, 1340–1351 (2010). 14. Carrino-Kyker, S. R., Coyle, K. P., Kluber, L. A. & Burke, D. J. Fungal and bacterial communities exhibit consistent responses to reversal of soil acidifcation and phosphorus limitation over time. Microorganisms 8, 1 (2019). 15. Sun, L., Chen, S., Chao, L. & Sun, T. Efects of fooding on changes in Eh, pH and speciation of cadmium and lead in contaminated soil. Bull. Environ. Contam. Toxicol. 79, 514–518 (2007). 16. Dordas, C. Role of nutrients in controlling plant diseases in sustainable agriculture: a review. Agron. Sustain. Dev. 28, 33–46 (2008). 17. Warren, S. L. J. H. Mineral nutrition of crops: fundamental mechanisms and implications. HortScience 39, 462–462 (2004). 18. Sharma, S., Duveiller, E., Basnet, R., Karki, C. B. & Sharma, R. Efect of potash fertilization on Helminthosporium leaf blight severity in wheat, and associated increases in grain yield and kernel weight. Field Crops Res. 93, 142–150 (2005). 19. Floch, C., Capowiez, Y. & Biology, S. Enzyme activities in apple orchard agroecosystems: how are they afected by management strategy and soil properties. Soil Biol. Biochem. 41, 61–68 (2009). 20. Aon, M. A. & Coloneri, A. C. II. Temporal and spatial evolution of enzymatic activities and physico-chemical properties in an agricultural soil. Appl. Soil Ecol. 18, 255–270 (2001). 21. Cai, Z. et al. Efects of the novel pyrimidynyloxybenzoic herbicide ZJ0273 on enzyme activities, microorganisms and its degrada- tion in Chinese soils. Environ. Sci. Pollut. Res. 22, 4425–4433 (2015). 22. Antonious, G. F. Impact of soil management and two botanical insecticides on urease and invertase activity. J. Environ. Sci. Health Part B Pestic. Food Contam. Agric. Wastes 38, 479–488 (2003). 23. Makoi, J. H. J. R. & Ndakidemi, P. A. Selected soil enzymes: examples of their potential roles in the ecosystem. Afr. J. Biotechnol. 7, 181–191 (2008). 24. Jian, S. et al. Soil extracellular enzyme activities, soil carbon and nitrogen storage under nitrogen fertilization: a meta-analysis. Soil Biol. Biochem. 101, 32–43 (2016). 25. Schutzendubel, A. & Polle, A. Plant responses to abiotic stresses: heavy metal-induced oxidative stress and protection by mycor- rhization. J. Exp. Bot. 53, 1351–1365 (2002). 26. Chao, A. Non-parametric estimation of the classes in a population. Scand. J. Stat. 11, 265–270 (1984). 27. Shannon, C. E. J. B. S. T. J. A mathematical theory of communication. Bell Syst. Tech. J. 27, 379–423 (1948). 28. Simpson, E. H. Measurement of diversity. Nature 163, 688 (1949).

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 10 Vol:.(1234567890) www.nature.com/scientificreports/

29. Farh, M. E., Kim, Y. J., Kim, Y. J. & Yang, D. C. Cylindrocarpon destructans/Ilyonectria radicicola-species complex: causative agent of ginseng root-rot disease and rusty symptoms. J. Ginseng Res. 42, 9–15 (2018). 30. Lombard, L., Der Merwe, N. A. V., Groenewald, J. Z. & Crous, P. W. Generic concepts in Nectriaceae. Stud. Mycol. 80, 189–245 (2015). 31. Martinezbarrera, O. Y. et al. Does Beauveria bassiana (Hypocreales: Cordycipitaceae) afect the survival and fecundity of the parasitoid Coptera haywardi (Hymenoptera: Diapriidae)?. Environ. Entomol. 48, 156–162 (2019). 32. Migiro, L. N., Maniania, N. K., Chabi-Olaye, A., Wanjoya, A. & Control, J. V. J. B. Efect of infection by Metarhizium anisopliae (Hypocreales: Clavicipitaceae) on the feeding and oviposition of the pea leafminer Liriomyza huidobrensis (Diptera: Agromyzidae) on diferent host plants. Biol. Control 56, 179–183 (2011). 33. Tardy, V. et al. Shifs in microbial diversity through land use intensity as drivers of carbon mineralization in soil. Soil Biol. Biochem. 90, 204–213 (2015). 34. Yang, L., Lu, X., Li, S. & Wu, B. First report of common bean (Phaseolus vulgaris) root rot caused by Plectosphaerella cucumerina in China. Plant Dis. 102, 1849–1849 (2018). 35. Baldeweg, F., Warncke, P., Fischer, D. & Gressler, M. J. O. L. Fungal biosurfactants from Mortierella alpina. Org. Lett. 21, 1444–1448 (2019). 36. Masinova, T., Yurkov, A. & Baldrian, P. J. F. E. Forest soil yeasts: decomposition potential and the utilization of carbon sources. Fungal Ecol. 34, 10–19 (2018). 37. Hu, H. et al. Fomesafen impacts bacterial communities and enzyme activities in the rhizosphere. Environ. Pollut. 253, 302–311 (2019). 38. Barriuso, J., Marín, S. & Mellado, R. P. Efect of the herbicide glyphosate on glyphosate-tolerant maize rhizobacterial communi- ties: a comparison with pre-emergency applied herbicide consisting of a combination of acetochlor and terbuthylazine. Environ. Microbiol. 12, 1021–1030 (2010). 39. Zhao, J. et al. Efects of organic–inorganic compound fertilizer with reduced chemical fertilizer application on crop yields, soil biological activity and bacterial community structure in a rice–wheat cropping system. Appl. Soil Ecol. 99, 1–12 (2016). 40. Davidova, I. A., Marks, C. R., & Sufita, J. M. Anaerobic hydrocarbon-degrading Deltaproteobacteria. Handbook of Hydrocarbon and Lipid Microbiology 207–243 (2019). 41. Wang, W. et al. Predatory Myxococcales are widely distributed in and closely correlated with the bacterial community structure of agricultural land. Appl. Soil Ecol. 146, 103365 (2020). 42. Yadav, S. et al. Cyanobacteria: role in agriculture, environmental sustainability, biotechnological potential and agroecological impact. In Plant-Microbe Interactions in Agro-Ecological Perspectives 257–277 (2017). 43. Rossi, F., Li, H., Liu, Y. & De Philippis, R. Cyanobacterial inoculation (cyanobacterisation): perspectives for the development of a standardized multifunctional technology for soil fertilization and desertifcation reversal. Earth Sci. Rev. 171, 28–43 (2017). 44. Muñoz-Rojas, M. et al. Efects of indigenous soil cyanobacteria on seed germination and seedling growth of arid species used in restoration. Plant Soil 429, 91–100 (2018). 45. Speirs, L. B. M., Rice, D. T. F., Petrovski, S. & Seviour, R. J. Te phylogeny, biodiversity, and ecology of the chlorofexi in activated sludge. Front. Microbiol. 10, 2015 (2019). 46. Whitman, T. et al. Dynamics of microbial community composition and soil organic carbon mineralization in soil following addi- tion of pyrogenic and fresh organic matter. ISME J. 10, 2918–2930 (2016). 47. Lladó, S. et al. Functional screening of abundant bacteria from acidic forest soil indicates the metabolic potential of Acidobacteria subdivision 1 for polysaccharide decomposition. Biol. Fertil. Soils 52, 251–260 (2016). 48. Bousset, L., Ermel, M., Soglonou, B. & Husson, O. Fungal growth is afected by and afects pH and redox potential (Eh) of the growth medium. bioRxiv 401182 (2018). 49. Mi, C. et al. Unveiling of dominant fungal pathogens associated with rusty root rot of Panax notoginseng based on multiple methods. Plant Dis. 101, 2046–2052 (2017). 50. Wang, Q. et al. Analysis of rhizosphere bacterial and fungal communities associated with rusty root disease of Panax ginseng. Appl. Soil Ecol. 138, 245–252 (2019). 51. Li, Z., Guo, S., Tian, S., Liu, Z. & Long, B. Study on the causes for ginseng red skin sickness occurred in albic bed soil. Acta Pedol. Sin. 34, 328–335 (1997). 52. Shi, J., Yuan, X., Lin, H., Yang, Y. & Li, Z. Y. Diferences in soil properties and bacterial communities between the rhizosphere and bulk soil and among diferent production areas of the medicinal plant Fritillaria thunbergii. Int. J. Int. J. Mol. Sci. 12, 3770–3785 (2011). 53. Xu, Y. et al. Bacterial communities in soybean rhizosphere in response to soil type, soybean genotype, and their growth stage. Soil Biol. Biochem. 41, 919–925 (2019). 54. Pei, G., Zhu, Y., Wen, J., Pei, Y. & Li, H. Vinegar residue supported nanoscale zero-valent iron: remediation of hexavalent chromium in soil. Environ. Pollut. 256, 113407 (2019). 55. Fawcett, J. K. Te semi-micro Kjeldahl method for the determination of nitrogen. J. Med. Lab Technol. 12, 1–22 (1954). 56. Li, Y. et al. Humic acid fertilizer improved soil properties and soil microbial diversity of continuous cropping peanut: a three-year experiment. Sci. Rep. 9, 12014 (2019). 57. Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37, 852–857 (2019). 58. Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17, 10–12 (2011). 59. Callahan, B. J. et al. DADA2: high-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581–583 (2016). 60. Katoh, K., Misawa, K., Kuma, K. & Miyata, T. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30, 3059–3066 (2002). 61. Price, M. N., Dehal, P. S. & Arkin, A. P. J. P. O. FastTree 2–approximately maximum-likelihood trees for large alignments. PLoS ONE 5, e9490 (2010). 62. Bokulich, N. A. et al. Optimizing taxonomic classifcation of marker-gene amplicon sequences with QIIME 2’s q2-feature-classifer plugin. Microbiome 6, 90 (2018). 63. Segata, N. et al. Metagenomic biomarker discovery and explanation. Genome Biol. 12, 1–18 (2011). Acknowledgements Te authors are grateful for the National key R&D program (Grant No. 2017YFC1702100). Author contributions L.Z. and X.B. designed the study. X.B. and H.Y. conducted the experiments. S.X., Y.Z., Y.X. and X.B. analysed the data. L.Z., S.X. and X.B. prepared the manuscript.

Competing interests Te authors declare no competing interests.

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 11 Vol.:(0123456789) www.nature.com/scientificreports/

Additional information Supplementary information is available for this paper at https​://doi.org/10.1038/s4159​8-020-71024​-8. Correspondence and requests for materials should be addressed to L.Z. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afliations. Open Access Tis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Te images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/.

© Te Author(s) 2020

Scientific Reports | (2020) 10:15756 | https://doi.org/10.1038/s41598-020-71024-8 12 Vol:.(1234567890)